From mboxrd@z Thu Jan  1 00:00:00 1970
Message-ID: <76d4d81b31c5601c83820bb37c83907b@proxima.alt.za>
To: 9fans@9fans.net
Date: Wed,  2 Jul 2008 06:52:07 +0200
From: lucio@proxima.alt.za
In-Reply-To: <970551641B57BC6070158BA7@[172.16.10.224]>
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
Subject: Re: [9fans] sad commentary
Topicbox-Message-UUID: d5832e60-ead3-11e9-9d60-3106f5b1d025

> It's fine, if you're fine with it ;-) Do you ever visit any AJAX enabled
> websites? Do you consider AJAX a superfluous technology? Do you switch to
> your "other OS" machine--or reboot your current machine--if and when you
> visit GMail's pages (at least to enable IMAP access for the first time)?
> What's your opinion on good ol' non-standard CSS? Won't you ever want to
> use one of these new "content delivery" systems, such as Microsoft
> Silverlight or Adobe Flash?

You're putting the cart before the horse.  Of course there is
"utility" computing out there and of course there is Microsoft to fill
that niche.  And Linux to follow in its footsteps.  But there is a
frightening prospect if you assume that utility computing is all that
computing is about, namely that only massive programming effort is
required to produce any sort of computing product.

Let me try this as a comparison.  Less than a hundred years ago,
Bugatti manufactured one motorvehicle a year, from scratch.  I'm not
sure how many persons were involved, the impression I have from
hearsay is that it was a single individual.  Today, you need the might
of the Chinese or Indian manufacturers to enter the motorvehicle
manufacturing business.  Or huge investment effort for the new
eco-friendly vehicles.

Utility computing is perfectly fine as long as it is balanced by
original development, but it is poisonous if it preclueds any original
participation.  Open Source is one form of rebellion, but it lacks the
robust foundations of sound program development.  Plan 9 is a much
smaller, better designed approach.  I'm sure we won't see Plan 9
deployed widely any time soon, it lacks the "utility" nature of the
contenders and I'm sorry to see that happening, but that is the nature
of the beast.  Had Plan 9 caught the imagination of the "masses", it
would have grown the same tumors as Linux, and that would have
defeated its nature.

Think Pascal: it is hardly the language of choice today, but the
principles it enshrines have totally altered the programming language
landscape.  C is the utility version, and C++ and Java its obvious
offsprings.  Alef has been abandoned and Limbo remains a very
specialised language, but they will also leave their mark.

So, I think this dicussion is based on a premise whose value is purely
emotional: we'd all be more comfortable if Plan 9 was widely accepted,
but there is no intellectual reason for it to be so.  Rob Pike says
the same thing in a nutshell, but in reality it is the philosophy
behind Plan 9 that needs spreading: careful design, generalised
objects, simplicity rather than bulk, etc.  Not Rio or Acme, Fossil or
Venti, but the environment in which they can thrive.  The environment
in which Mozilla is difficult to create so that simpler solutions can
be sought.

++L